ABSTRACT Alexander disease (AxD) is a primary disease of astrocytes caused by mutations in the gfap gene. How AxD mutations lead to protein aggregation and astrocyte dysfunction as well as how mutant GFAP-expressing astrocytes result in neuronal degeneration remain unknown. Evolutionarily, astrocytes play increasingly more important and complex roles in human brain functions. Hence, the ability to directly examine AxD patients' astrocytes will not only complement the existing models but also reveal potential unique aspects of human astrocytes in AxD pathogenesis. We have built induced pluripotent stem cells (iPSCs) from AxD patients with three different mutations through close collaboration with the Messing lab (project 3). We have also developed a reproducible strategy to guide hPSCs to enriched astrocytes. Our preliminary study revealed the presence of RFs in AxD patients' astrocytes in culture and following transplantation into the mouse brain, highlighting the recapitulation of key AxD pathology in our iPSC system. However, astrocytes derived from AxD patients exhibit comparable GFAP levels as non-AxD individuals, suggesting that GFAP may not increase at early stages of AxD and a simple increase in GFAP protein might not be the major trigger for RF formation and astrocyte dysfunction in human cells as proposed for model animals. Our ability to produce enriched astrocytes from AxD patients has also led to the discovery that AxD astrocytes have altered ratio of GFAP-d/a isoforms, which has led to the ?rediscovery? of a similar ratio change in transgenic mice. We will use advanced gene editing technology to determine if change of GFAP isoforms mediates the effects of gfap mutations on protein aggregation and/or astrocyte dysfunction. By neuron-astrocyte co-culture and neural transplantation, we will determine if and how AxD astrocytes cause neuronal degeneration. As part of the program project, we will validate modifiers of AxD learned from animal models in our AxD patients' astrocytes/neurons, setting up the foundation for future translation.